Instead of selling software to traditional industries, a more defensible approach is to build vertically integrated companies. This involves acquiring or starting a business in a non-sexy industry (e.g., a law firm, hospital) and rebuilding its entire operational stack with AI at its core, something a pure software vendor cannot do.

As AI infrastructure giants become government-backed utilities, their investment appeal diminishes like banks after 2008. The next wave of value creation will come from stagnant, existing businesses that adopt AI to unlock new margins, leveraging their established brands and distribution channels rather than building new rails from scratch.

Unlike traditional software development, AI-native founders avoid long-term, deterministic roadmaps. They recognize that AI capabilities change so rapidly that the most effective strategy is to maximize what's possible *now* with fast iteration cycles, rather than planning for a speculative future.

Don't just sprinkle AI features onto your existing product ('AI at the edge'). Transformative companies rethink workflows and shrink their old codebase, making the LLM a core part of the solution. This is about re-architecting the solution from the ground up, not just enhancing it.

Incumbent companies are slowed by the need to retrofit AI into existing processes and tribal knowledge. AI-native startups, however, can build their entire operational model around agent-based, prompt-driven workflows from day one, creating a structural advantage that is difficult for larger companies to copy.

Startups like Cognition Labs find their edge not by competing on pre-training large models, but by mastering post-training. They build specialized reinforcement learning environments that teach models specific, real-world workflows (e.g., using Datadog for debugging), creating a defensible niche that larger players overlook.

Small firms can outmaneuver large corporations in the AI era by embracing rapid, low-cost experimentation. While enterprises spend millions on specialized PhDs for single use cases, agile companies constantly test new models, learn from failures, and deploy what works to dominate their market.

For incumbent software companies, an existing customer base is a double-edged sword. While it provides a distribution channel for new AI products, it also acts as "cement shoes." The technical debt and feature obligations to thousands of pre-AI customers can consume all engineering resources, preventing them from competing effectively with nimble, AI-native startups.

Most current AI tools are skeuomorphic—they just perform old tasks more efficiently. The real transformation will come from "AI-native" applications that create entirely new business models, just as Uber was an "iPhone-native" concept unimaginable before its time. The biggest winners will use AI to become the industry, not just sell to it.